Communication disorders affect millions of people. Understanding the neural bases of vocal learning will enable early diagnosis and effective treatment of these diseases. There are a few nonhuman vocal learners, of which the songbirds offer the best-characterized model in terms of physiology and behavior. As with humans, the zebra finch songbird learns its song in two phases: a sensory phase during which the song of an adult tutor is memorized and a sensorimotor phase during which the birds own vocalizations are shaped through auditory feedback to match the tutor song. The shaping of the vocalization requires that the comparison between auditory feedback and the memory of the tutor song sculpt the song motor control circuitry. We recently found that the tutor song selectively activates a key nucleus of the song premotor pathway (Nick & Konishi, 2005a). This selective activation occurs only during waking and only during the period of development when the tutor song memory is used to shape vocalizations. This suggests that the comparison of auditory feedback with the tutor song memory generates an instructive matching signal that is relayed to the premotor nucleus. There are 4 specific hypotheses: (1) responses of individual neurons convey the degree of similarity between stimuli and the tutor song memory; (2) the matching signal occurs during singing; (3) the matching signal is relayed to the basal ganglia; and (4) the mechanism that transforms the matching signal into behavioral change involves sustained neural activity in the song system that enables temporal overlap of motor command and sensory feedback and subsequent activity-dependent plasticity. The study will utilize three powerful techniques in awake juveniles: multi-electrode recording, which enables the stable assessment of the activity of many single neurons, antidromic stimulation, which enables the identification of individual neurons, and long-term population recordings. The ultimate goal is to use the matching signal to illuminate the role of memory and sensation in shaping vocal behavior. The candidate, Dr. Teresa A. Nick, is uniquely qualified to execute these experiments. She has received training on (1) the development of neurons and circuits from Drs. Thomas Carew and Leonard Kaczmarek (Yale); (2) extrinsic modulation of neuronal development from Dr. Angeles Ribera (Univ. Colorado); and (3) the development, state-dependent modulation, and learning of birdsong from Dr. Masakazu Konishi (Caltech). She has published extensively on neural development and has discovered the first evidence for a template-matching signal. She has already applied several novel techniques to the song system and developed a new method combining multi-electrode and antidromic techniques. The University of Minnesota provides the ideal environment in which to pursue these experiments, due to strengths in auditory processing, multi-electrode techniques, and antidromic methods.